© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 43949
    Prev
    Next
    INDEX
    Explanations

    Jr

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron detects the little reference‐style metadata tokens—single‐letter author initials, volume/page numbers, and the hyphen separators—typical of academic citation entries.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     our
    -1.77
     like
    -1.60
     your
    -1.52
     consultez
    -1.47
    いただきたい
    -1.45
     to
    -1.43
     them
    -1.39
     solicitudes
    -1.38
    курс
    -1.38
     ajustes
    -1.36
    POSITIVE LOGITS
     некоторых
    1.46
    ianum
    1.45
     wich
    1.38
     who
    1.38
     åren
    1.34
    武道
    1.33
     gmbh
    1.33
     \
    1.29
    Выводы
    1.29
    if
    1.28
    Activations Density 0.013%

    No Known Activations